Membrane proteins (MPs) constitute a special set of problems concerning expression in a functional form, and purification to a pure, homogeneous and stable entity suitable for structural approaches using both cryo-EM (Core 3) and X-ray methods (Core 6). Less than 1% of the entries in the protein structure database are membrane proteins and most of these are bacterial proteins, reflecting the much greater difficulties of suitable expression, purification and crystallization of membrane proteins, and particularly eukaryotic membrane proteins. Among the accessory proteins in HIV is the membrane protein Vpu. Vpu interacts with several host soluble and membrane proteins including two well-characterized targets for degradation via ubiquitination pathways, these being the receptor for HIV, CD4, and BST-2 (tetherin). Mutation of the gene, selection of constructs, and expression of Vpu, and formation of complexes with both soluble and membrane proteins are therefore a high priority. Vpu also forms a pentameric viroporin structure that we prepare for structure analysis and determination in Project 3. The accessory protein Nef, together with the clathrin adaptor AP-2 and other factors, form an interaction network with host 10-crossing integral membrane proteins SERINC3, or SERINC5 to target these restriction factors for destruction. Expression of SERINC3 and 5 are a priority for the goals of Project 4, which are to define this interaction network at atomic resolution. The core will screen a variety of species and constructs to obtain the quantity and quality of protein needed for structure analysis. We have developed a general method to prioritize, express, purify and crystallize integral membrane proteins. We will express, purify, thermodynamically stabilize and produce membrane proteins (MPs) suitable for cryo-EM and/or X-ray diffraction of complexes. Prioritized complexes will be assayed as to homogeneity, purity and stability. Purification will be optimized for structure determination by cryo-EM imaging, which is particularly well suited to complexes of relatively large size, and for crystallization. Crystallization will use detergent-lipid screens, bicelles, nanodiscs, cubic and sponge phase crystallizations (Core 6).